Study 103 Correlational.PNG
Celebration Thread! Biogen is going to the FDA with Aducan.
Re: Celebration Thread! Biogen is going to the FDA with Aducan.
You do not have the required permissions to view the files attached to this post.
Re: Celebration Thread! Biogen is going to the FDA with Aducan.
When considering the aducan correlation to amyloid this one is always a good chestnut to refer to. It is quite striking how powerfully those with more amyloid clearance did versus those less. This is based on a nice sample size and as we see those with the most amyloid clearance over 1 year had stabilized MMSE. Also notice the error bars-- nice and small.
What might explain the conflicting results for 103 and 301 and 301? The patient population was less advanced in the phase 3s with many of the placebo not even advancing.
Figure 16 above includes regression lines that appear to be so sensitive that even single points on the edges drives the lines to extremes.
What might explain the conflicting results for 103 and 301 and 301? The patient population was less advanced in the phase 3s with many of the placebo not even advancing.
Figure 16 above includes regression lines that appear to be so sensitive that even single points on the edges drives the lines to extremes.
Last edited by J11 on Sat Dec 12, 2020 11:54 am, edited 1 time in total.
Re: Celebration Thread! Biogen is going to the FDA with Aducan.
On page 254, they discuss the results conditioned on APOE epsilon 4 genotype. The 302 non-carriers were worse than placebo on MMSE, epsilon 4+ was in the wrong direction in 301 etc. . MMSE is not a good instrument to consider for early stage AD patients. It is probably best to briefly mention it at the beginning of the discussion and then move on to better tests. Why did not Appendix 2 also include the other tests (e.g., ADAS-cog and ADCS-ADL? It would also be of interest to see how the CDRsb of the APOE sub-groups were changing through time. It is known that the top line of 301 did not hit its number. Simply repeating this without giving some sense of the temporal changes even from the December 2019 interim futility analysis and then the termination announcement on March 21,2019 seems highly deceptive. The p=0.6971 quoted for the APOE+ underwent rapid change over a short period of time. Without providing more of the dataset, narratives can be constructed that do not conform to a true interpretation of the overall numbers. This was seen with the 302 "high dose" subgroup that often was not high dose. With the very large treatment effects that have be seen with aducan only such fudges can disguise the impressive results from the trials.
The issue raised about ARIA unblinding was covered earlier in the document. Sensitivity analysis found that the results were robust to ARIA. As also needs to be remembered, ARIA at some level is a sign of amyloid clearance. Some of these patients actually achieved better results with this clearance providing evidence connecting amyloid removal and cognition. Yet, over the shorter term ARIA can also cause potential cognitive impairment.
My interpretation of the Appendix 2 Executive Summary is that it is presenting weak counter-arguments against aducan and along the way scores several own goals. Are these really the strongest opposing arguments they have? From here on out it is going to be talking about small sub-groups and not acknowledging the lack of high dosing in the high dose group?
The issue raised about ARIA unblinding was covered earlier in the document. Sensitivity analysis found that the results were robust to ARIA. As also needs to be remembered, ARIA at some level is a sign of amyloid clearance. Some of these patients actually achieved better results with this clearance providing evidence connecting amyloid removal and cognition. Yet, over the shorter term ARIA can also cause potential cognitive impairment.
My interpretation of the Appendix 2 Executive Summary is that it is presenting weak counter-arguments against aducan and along the way scores several own goals. Are these really the strongest opposing arguments they have? From here on out it is going to be talking about small sub-groups and not acknowledging the lack of high dosing in the high dose group?
Re: Celebration Thread! Biogen is going to the FDA with Aducan.
That was the executive summary, and these perspectives will be further elaborated upon in the statistical report. Closer scrutiny will be given as they are encountered.
You do not have the required permissions to view the files attached to this post.
Re: Celebration Thread! Biogen is going to the FDA with Aducan.
The above is the location of the data files included for the response in Appendix 2. It would be helpful to publicly disclose these files (if possible) in order to better understand the aducan clinical trials.
Engage and Emerge were followed by an optional dose-blinded long-term extension (LTE) period of up to 5 years. Clarification about the current state of knowledge about the Engage and Emerge (i.e., data collected subsequent to the termination of the trials) would be extremely helpful. One can speculate that the reason why no additional clinical trial results were required is that they were unnecessary. All that might be required for all discussion to end about aducan efficacy is another cognitive readout for those on long term extension.
Figure 7 which showed the number of doses received on high dose 302 showed all of these patients at randomization and the number of high doses that they received. This was up to around ~24 September 2017. Yet, the transfer to PV4 along with many patients receiving up to 14 high doses begins ~January 2017. ~10 patients per month of these high number of high dose aducan begin to accumulate starting from January 2017. Potentially what this means is that ~10 high dose patients could have 10 months of additional high dose aducan treatment, another ~10 with 9 months and so on for 302. Possibly similar numbers might be available for the 301 study. It mentioned that it was dose-blinded so perhaps everyone would not be on high- dose, it also is not clear whether there was ongoing cognitive monitoring.
10 months on additional aducan treatment? Potentially 10 more aducan 10 mg/kg doses? This is not even worth a bitter brawl. That much additional dosing should show overwhelming efficacy. Going from 13 doses to 14 doses of high dose increased the benefit over placebo from 24% to 30% (i.e,., a 25% change). What happens when you go from 14 doses to 24 doses? I do not want to avoid that question as it is an opportunity to demonstrate understanding. My ~ guess would be that at a minimum this would mean a 50% benefit over placebo (with reasonable historical placebos if needed).
This information does not appear to have been disclosed in the FDA Briefing Document. For the sake of transparency this information should be disclosed.
Engage and Emerge were followed by an optional dose-blinded long-term extension (LTE) period of up to 5 years. Clarification about the current state of knowledge about the Engage and Emerge (i.e., data collected subsequent to the termination of the trials) would be extremely helpful. One can speculate that the reason why no additional clinical trial results were required is that they were unnecessary. All that might be required for all discussion to end about aducan efficacy is another cognitive readout for those on long term extension.
Figure 7 which showed the number of doses received on high dose 302 showed all of these patients at randomization and the number of high doses that they received. This was up to around ~24 September 2017. Yet, the transfer to PV4 along with many patients receiving up to 14 high doses begins ~January 2017. ~10 patients per month of these high number of high dose aducan begin to accumulate starting from January 2017. Potentially what this means is that ~10 high dose patients could have 10 months of additional high dose aducan treatment, another ~10 with 9 months and so on for 302. Possibly similar numbers might be available for the 301 study. It mentioned that it was dose-blinded so perhaps everyone would not be on high- dose, it also is not clear whether there was ongoing cognitive monitoring.
10 months on additional aducan treatment? Potentially 10 more aducan 10 mg/kg doses? This is not even worth a bitter brawl. That much additional dosing should show overwhelming efficacy. Going from 13 doses to 14 doses of high dose increased the benefit over placebo from 24% to 30% (i.e,., a 25% change). What happens when you go from 14 doses to 24 doses? I do not want to avoid that question as it is an opportunity to demonstrate understanding. My ~ guess would be that at a minimum this would mean a 50% benefit over placebo (with reasonable historical placebos if needed).
This information does not appear to have been disclosed in the FDA Briefing Document. For the sake of transparency this information should be disclosed.
Re: Celebration Thread! Biogen is going to the FDA with Aducan.
You do not have the required permissions to view the files attached to this post.
Re: Celebration Thread! Biogen is going to the FDA with Aducan.
You do not have the required permissions to view the files attached to this post.
Re: Celebration Thread! Biogen is going to the FDA with Aducan.
This is returning to the initial part of the Briefing Document. What the above figures appear to show is that there was a large amount of post-trial dosing in the long term extensions. How this could have been possibly overlooked in the advisory meeting is very unclear.
All of this extra dosing makes essentially all additional argument moot.
The Briefing Document only refers to the on study dosing. Apparently, the long-term extension had blinded dosing, so an aspect of blindedness remained.
It is not entirely clear whether the 10 mg/kg on page 5 of 6 on page 107 refers to the study arm of the patients or the dosing received on long term extension.
However, 792 patients are reported to have received at least one 10mg/kg dose on long term extension,
A total of 1623 patients received at least one aducan dose on long-term extension.
Page 2 of 6 notes that 582 patients on 10 mg/kg received treatment for greater than or equal to 18 months, with 319 on treatment for more than 2 years and 65 more more than 3 years. At the 2 year mark patients would receive 20 high doses and at 3 years 32 high doses of aducan.
Not only was the futility analysis futile, but so too is arguing about the effectiveness of aducan. The 319 patients who received 20 high doses would provide substantial information about efficacy. Now that the COVID vaccine is expected to start rolling out on Monday, we will soon have thousands of patients back dosed on aducan. With this group they will act as their own control which will help avoid the rapid progressor problem.
Page 2 also tells us that patients were on 10 mg/kg treatment for up to 3.5 years (38 doses). also of interest is that they note that these were continuous doses so interruptions due to ARIA etc.. might be unreported here.
All of this extra dosing makes essentially all additional argument moot.
The Briefing Document only refers to the on study dosing. Apparently, the long-term extension had blinded dosing, so an aspect of blindedness remained.
It is not entirely clear whether the 10 mg/kg on page 5 of 6 on page 107 refers to the study arm of the patients or the dosing received on long term extension.
However, 792 patients are reported to have received at least one 10mg/kg dose on long term extension,
A total of 1623 patients received at least one aducan dose on long-term extension.
Page 2 of 6 notes that 582 patients on 10 mg/kg received treatment for greater than or equal to 18 months, with 319 on treatment for more than 2 years and 65 more more than 3 years. At the 2 year mark patients would receive 20 high doses and at 3 years 32 high doses of aducan.
Not only was the futility analysis futile, but so too is arguing about the effectiveness of aducan. The 319 patients who received 20 high doses would provide substantial information about efficacy. Now that the COVID vaccine is expected to start rolling out on Monday, we will soon have thousands of patients back dosed on aducan. With this group they will act as their own control which will help avoid the rapid progressor problem.
Page 2 also tells us that patients were on 10 mg/kg treatment for up to 3.5 years (38 doses). also of interest is that they note that these were continuous doses so interruptions due to ARIA etc.. might be unreported here.
Re: Celebration Thread! Biogen is going to the FDA with Aducan.
On page 260, MMSE is shown as the top ranked secondary endpoint in the rank prioritization. It is very unclear why the MMSE would be given this rank as it is not known to possess strong psychometric properties especially in early AD populations. The other secondary tests reported strong results in the high 302 trial.
On page 261, it is noted that the low dose acted as a gatekeeper and if statistical significance was not reached then the other secondaries should not be considered due to multiple comparisons. As is shown later combining the 301 and 302 low doses did actually achieve statistical significance. The low dose 301 + 302 numbers were CDRsb p= 0.0513, ADAS-cog p= 0.0764, ADCS-ADL p= 0.0347, and MMSE p= 0.8346 (page 23).
I realize that you want to stay with the plan and you really should stay with the plan, though sometimes plans change. The idea that you should toss the ADCS-ADL p= 0.0006 because the low dose did not reach significance does not seem reasonable. The low dose did show that it trended away from placebo which was really what it was intended to do. This represents more evidence that a dose effect does exist. If needed, they could pay an alpha penalty and leave it at that. Rigidly adhering to the statistical plan is not overly helpful. There were a great many aspects of the trials that were unknown at the start; even the high dosing that should be used which was eventually increased to 10 mg/kg with evolving evidence from the 103 study. All of this confusion arose for the very reason that aducan was so extremely promising that phase 2 trials were thought unnecessary. In this context, statistical analysis only can tell you so much and such analysis should not attain more importance than the background noise is generating.
On page 261, it is noted that the low dose acted as a gatekeeper and if statistical significance was not reached then the other secondaries should not be considered due to multiple comparisons. As is shown later combining the 301 and 302 low doses did actually achieve statistical significance. The low dose 301 + 302 numbers were CDRsb p= 0.0513, ADAS-cog p= 0.0764, ADCS-ADL p= 0.0347, and MMSE p= 0.8346 (page 23).
I realize that you want to stay with the plan and you really should stay with the plan, though sometimes plans change. The idea that you should toss the ADCS-ADL p= 0.0006 because the low dose did not reach significance does not seem reasonable. The low dose did show that it trended away from placebo which was really what it was intended to do. This represents more evidence that a dose effect does exist. If needed, they could pay an alpha penalty and leave it at that. Rigidly adhering to the statistical plan is not overly helpful. There were a great many aspects of the trials that were unknown at the start; even the high dosing that should be used which was eventually increased to 10 mg/kg with evolving evidence from the 103 study. All of this confusion arose for the very reason that aducan was so extremely promising that phase 2 trials were thought unnecessary. In this context, statistical analysis only can tell you so much and such analysis should not attain more importance than the background noise is generating.
Re: Celebration Thread! Biogen is going to the FDA with Aducan.
You do not have the required permissions to view the files attached to this post.
